We describe scoring metrics for learning Bayesian networks from a combination of user knowledge and statistical data. We identify two important properties of metrics, which we call event equivalence and parameter modularity. These properties have been mostly ignored, but when combined, greatly simplify the encoding of a user’s prior knowledge. In particular, a user can express his knowledge—for the most part—as a single prior Bayesian network for the domain. 1

Abstract — In this paper the problem of dynamic selfreconfiguration of a class of modular robotic systems referred to as metamorphic systems is examined. A metamorphic robotic system is a collection of mechatronic modules, each of which has the ability to connect, disconnect, and climb over adjacent modules. We examine the near-optimal reconfiguration of a metamorphic robot from an arbitrary initial configuration to a desired final configuration. Concepts of distance between metamorphic robot configurations are defined, and shown to satisfy the formal properties of a metric. These metrics, called configuration metrics, are then applied to the automatic self-reconfiguration of metamorphic systems in the case when one module is allowed to move at a time. There is no simple method for computing the optimal sequence of moves required to reconfigure. As a result, heuristics which can give a near optimal solution must be used. We use the technique of Simulated Annealing to drive the reconfiguration process with configuration metrics as cost functions. The relative performance of simulated annealing with different cost functions is compared and the usefulness of the metrics developed in this paper is demonstrated. Index Terms—Metrics, optimal assignment, self-reconfigurable robots, simulated annealing.

Abstract not found

Abstract-- In basic genetic algorithm (GA) applications, the fitness of a solution takes a value that is certain and unchanging. There are two classes of problem for which this formulation is insufficient. The first consists of ongoing searches for better solutions in a nonstationary environment, where the expected fitness of a solution changes with time in unpredictable ways. The second class consists of applications in which fitness evaluations are corrupted by noise. For problems belonging to either or both of these classes, the estimated fitness of a solution will have an associated uncertainty. Both the uncertainty due to environmental changes (process noise) and the uncertainty due to noisy evaluations (observation noise) can be reduced, at least temporarily, by re-evaluating existing solutions. The Kalman formulation provides a well-developed formal mechanism for treating uncertainty within the GA framework. It provides the mechanics for determining the estimated fitness and uncertainty when a new solution is generated and evaluated for the first time. It also provides the mechanics for updating the estimated fitness and uncertainty after an existing solution is re-evaluated, and for increasing the uncertainty with the passage of time. A Kalman-extended genetic algorithm (KGA) is developed to determine when to generate a new individual, when to re-evaluate an existing individual, and which one to re-evaluate. This KGA is applied to the problem of maintaining a network configuration with minimized message

Abstract not found

Accepted for publication by

Abstract not found

Maps are vital tools for most government agencies and consumers. However, their manual generation and updating is tedious, time consuming, and expensive. To address these concerns, we are developing automated techniques. In this paper, we restrict our attention to residential regions. These regions provide a challenge, testing the current limits of automated image analysis. Such regions are also typically areas of rapid growth and development and, therefore, are of interest from the applications perspective. In previous studies, we introduced statistical measures to extract these kinds of regions from satellite images [in: Proceedings of the International Conference on Pattern Recognition, vol. 1, 2002, p. 127, IEEE Trans. GeoRS (2003), IEEE Trans. PAMI]. As the next step toward automatic map generation, here we introduce a novel system to detect houses and street networks in IKONOS multispectral images. These images have one meter panchromatic resolution with 4 m resolution in the spectral bands. Our system consists of four major components: multispectral analysis to detect cultural activity, segmentation of regions of possible human activity (based on the surface material), decomposition of the segmented images, and graph theoretical algorithms over the decompositions to extract the street network and to detect houses. We tested our system on a large and diverse data set. Our results indicate the usefulness of our system in detecting houses and street networks, hence generating automated maps.

This paper deals with the problem of aligning data and computations when mapping affine loop nests onto Distributed Memory Parallel Computers (DMPCs). We formulate the problem by introducing a new graph, the access graph, to model affine communications (with rectangular access matrices) more adequately than with the previously introduced tool, the communication graph. We show that maximizing the number of local communications in the access graph is a NPcomplete problem in the strong sense and we present several heuristics based upon the access graph for mapping affine loop nests onto DMPCs. 1 Introduction This paper deals with the problem of mapping affine loop nests onto Distributed Memory Parallel Computers (DMPCs). Because the communication is very expensive in DMPCs, how to distribute data arrays and computations to processors is a key factor to performance. The computations described in this paper are general non-perfect loop nests (or multiple loop nests) with uniform or affine...

Cognitive radio networks have been proposed as a solution to both spectrum inefficiency and spectrum scarcity problems. However, they face to a unique challenge based on the fluctuating nature of heterogeneous spectrum bands as well as the diverse service requirements of various applications. In this paper, a spectrum decision framework is proposed to determine a set of spectrum bands by considering the application requirements as well as the dynamic nature of spectrum bands. To this end, first, each spectrum is characterized by jointly considering primary user activity and spectrum sensing operations. Based on this, a minimum variancebased spectrum decision is proposed for real-time applications, which minimizes the capacity variance of the decided spectrum bands subject to the capacity constraints. For best-effort applications, a maximum capacity-based spectrum decision is proposed where spectrum bands are decided to maximize the total network capacity. Moreover, a dynamic resource management scheme is developed to coordinate the spectrum decision adaptively dependent on the time-varying cognitive radio network capacity. Simulation results show that the proposed methods provide efficient bandwidth utilization while satisfying service requirements.